With the advancement in artificial intelligence (AI) and machine learning (ML) techniques, researchers are striving towards employing these techniques for advancing clinical practice. One of the key objectives in healthcare is the early detection and prediction of disease to timely provide preventive interventions. This is especially the case for epilepsy, which is characterized by recurrent and unpredictable seizures. Patients can be relieved from the adverse consequences of epileptic seizures if it could somehow be predicted in advance. Despite decades of research, seizure prediction remains an unsolved problem. This is likely to remain at least partly because of the inadequate amount of data to resolve the problem. There have been exciting new developments in ML-based algorithms that have the potential to deliver a paradigm shift in the early and accurate prediction of epileptic seizures. Here we provide a comprehensive review of state-of-the-art ML techniques in early prediction of seizures using EEG signals. We will identify the gaps, challenges, and pitfalls in the current research and recommend future directions.

ObjectiveTo investigate the hypothesis that patients diagnosed with psychogenic nonepileptic seizures (PNES) on video-EEG monitoring (VEM) have increased mortality by comparison to the general population.MethodsThis retrospective cohort study included patients evaluated in VEM units of 3 tertiary hospitals in Melbourne, Australia, between January 1, 1995, and December 31, 2015. Diagnosis was based on consensus opinion of experienced epileptologists and neuropsychiatrists at each hospital. Mortality was determined in patients diagnosed with PNES, epilepsy, or both conditions by linkage to the Australian National Death Index. Lifetime history of psychiatric disorders in PNES was determined from formal neuropsychiatric reports.ResultsA total of 5,508 patients underwent VEM. A total of 674 (12.2%) were diagnosed with PNES, 3064 (55.6%) with epilepsy, 175 (3.2%) with both conditions, and 1,595 (29.0%) received other diagnoses or had no diagnosis made. The standardized mortality ratio (SMR) of patients diagnosed with PNES was 2.5 (95% confidence interval [CI] 2.0–3.3). Those younger than 30 had an 8-fold higher risk of death (95% CI 3.4–19.8). Direct comparison revealed no significant difference in mortality rate between diagnostic groups. Among deaths in patients diagnosed with PNES (n = 55), external causes contributed 18%, with 20% of deaths in those younger than 50 years attributed to suicide, and “epilepsy” was recorded as the cause of death in 24%.ConclusionsPatients diagnosed with PNES have a SMR 2.5 times above the general population, dying at a rate comparable to those with drug-resistant epilepsy. This emphasizes the importance of prompt diagnosis, identification of risk factors, and implementation of appropriate strategies to prevent potential avoidable deaths.